IT ONLY TAKES A FEW HOURS to have an implantable cardioverter defibrillator (ICD) placed in your chest. During the procedure, which may be performed under local anesthesia and a sedative, the programmable, shock-generating device is tucked near your collarbone with electrical leads snaked through a vein to reach your heart’s right ventricle. An ICD can deliver either paced beats or electric shocks to control irregular heartbeats, including those that might lead to sudden cardiac death.

Yet while having an ICD in your chest may save your life, it also complicates it. You won’t know when or whether you’ll receive a shock, and the sudden jolt can be painful. Some people with ICDs have reported anxiety or depression.

Is it worth it? That may depend on your health, your temperament and how an ICD compares with alternative treatment options—considerations you might expect to go over carefully with your doctor. But a study published this spring in JAMA Internal Medicine revealed that such conversations are often cursory. Researchers asked 41 patients what they recalled about their communication with their doctors. Eighty percent didn’t remember discussing long-term consequences or post-surgery risks, and some weren’t aware that the devices occasionally malfunctioned. Many patients felt the decision had been rushed, says lead author Paul Hauptman, a cardiologist at Saint Louis University.

For a second part of the study, the researchers recruited 11 cardiologists to evaluate two “patients” (actors following a script) who qualified for an ICD. Most physicians focused on explaining the devices and the surgery, while few mentioned quality-of-life issues or long-term consequences. Many laced their explanations with medical jargon.

In a commentary on the study, Grace Lin, a physician and health policy researcher at the University of California, San Francisco, wrote that it “adds to the growing body of evidence that there is inadequate communication and little shared decision making occurring for situations such as ICD surgery.”

Increasingly, there’s a notion that medical decisions should be shared, particularly when patients may have more than one reasonable option. In “shared decision making,” as it has come to be called, physician and patient work together to select tests or therapies, considering not only medical evidence but also what patients want. Some might choose to have an ICD implanted if there is even a small chance they would benefit, while others could prefer to avoid the anxiety of having the device in their chests. Still others might have trouble understanding the full implications of the choice no matter how well it was explained to them. But the point is to make every effort to give the people who bear the consequences of health decisions a say in making those choices.

The idea that it’s important for patients to be involved in decision making is hardly new, says Karen Sepucha, director of the Health Decision Sciences Center at Massachusetts General Hospital. But there’s increasing evidence that it helps patients become better informed about their conditions and the risks and benefits of treatment options. It may also limit overuse of medical services and may result in patients’ being more inclined to stick with their therapy. At the same time, physicians face obstacles—liability fears, time and financial pressures—that may limit patients’ participation. For shared decision making to be successful, those concerns need to be addressed, and physicians and patients must rethink their traditional roles.

SOME MEDICAL CHOICES ARE CLEAR: A patient with acute bacterial meningitis needs antibiotics. But decisions about elective surgery, whether to take a medication for a chronic condition and whether to get a cancer screening test are less obvious. “Most of the time the evidence is not black or white,” says France Légaré, a researcher at Université Laval in Quebec.

Consider statins. Based on data from clinical trials, if 200 people with a history of heart disease or heart attack take the cholesterol-lowering drugs for five years, between two and 10 will avoid a heart attack. Understanding those fairly slim odds of protection, patients may well want to factor in other benefits, side effects, risks and the financial cost. Légaré says that when her team tested tools that presented positives and negatives, patients embraced them.

Patients, used to being told what they need, tend to assume that if a doctor recommends a treatment, it will help. In Hauptman’s ICD study, for example, patients were asked to estimate how many people, out of 100 who received the device, would likely benefit. The average of their guesses: 88. Yet implanting an ICD for patients at risk of sudden death because of an arrhythmia—the indication in this study—will probably save just seven or eight of each 100 who get the device. (Of course, there’s no guarantee that the family of a patient who doesn’t get an ICD and subsequently dies won’t sue the doctor.)

Even when there may be few clearly right or wrong choices, physicians may not take into account the weight their patients could give to various considerations. And doctors may routinely follow treatment protocols that were never intended to be hard and fast rules. “Guidelines are meant to support decision making, not be applied blindly to every patient,” Lin says.

A national survey of more than 3,000 adults age 40 and older, led by researchers at the University of Michigan, asked whether respondents had talked with a doctor in the preceding two years about one of nine decisions considered to be “preference sensitive”—screening for breast, colon or prostate cancer; treatment for depression, high blood pressure or elevated cholesterol; and surgery for back pain, cataracts or joint replacement. Patients reported their doctors had mostly pointed them toward a single option, and recommendations generally called for doing something. That could reflect pressures on physicians to intervene, including the threat of lawsuits and performance quality metrics based on screening or treatment rates.


ONE WAY TO HELP PATIENTS IS to give them brochures or electronic communications that lay out the pros and cons of a treatment. For example, patients who are scheduled to see Andrew A. Freiberg, chief of MGH’s knee and hip replacement service, are mailed just such a “decision aid”—a DVD that explains osteoarthritis and potential treatments. Patients learn that there are numerous options—including over the counter medications, weight loss, physical therapy, injections, or simple observation—and that each has risks and benefits. The DVD includes videos of real patients describing their choices.

“I was skeptical, but this has been a real asset,” Freiberg says. “Most people don’t initially understand what it means to have a major procedure.” The decision aid helps patients prepare for whatever treatment they choose. “These guides help elevate patients’ knowledge bases.”

Such tools have emerged as an important part of shared decision making. The goal, says Leigh Simmons, a primary care physician who co-directs the Shared Decision Making Program at MGH, is to help patients be on equal footing with physicians. The hospital is one of 10 primary care sites funded by the Informed Medical Decisions Foundation, a Boston nonprofit, to test the decision aids it has developed.

Joint replacement and other elective procedures are ideal for decision aids because there’s usually no rush to make a choice. But last year, a team at the Mayo Clinic in Rochester, Minn., showed that a similar process could work for patients who arrived at the emergency department complaining of acute chest pain. Normally, even those at low risk of a heart attack would likely be admitted for precautionary cardiac tests. For this study, however, some patients got a simple decision aid with a graphical representation of their risk of having a “pre-heart-attack” diagnosis during the next 45 days—98 human figures in blue who, statistically speaking, would not receive the diagnosis, and two in red who would. Patients could choose to be admitted for testing, to make an appointment with a heart doctor within a 24-to-72-hour period, to see a physician on their own or to have an emergency department doctor make the decision for them. Patients who saw the decision aid were more likely than those who didn’t to involve themselves in the choice and were less likely to seek immediate admission.

A 2011 analysis of 86 randomized clinical trials by Cochrane Reviews concluded that decision aids make patients better informed, improve communication with their doctors and ?increase participation in decisions about their care. Patients using decision aids also tend to be less likely to have surgery and more likely to limit other invasive treatments. A study published in Health Affairs last year found that giving decision aids to members of Group Health, a large system in Washington State and northern Idaho, resulted in 26% fewer hip replacements, 38% fewer knee replacements and a reduction in costs of 12% to 21% during a six-month period.

David Arterburn, lead author of the study, says that the decisions of patients with osteoarthritis hinge on their symptoms, how much those matter to them and whether they feel the time is right for surgery or they would prefer to try something less invasive first. This and similar studies have grabbed the attention of policymakers who believe decision aids could reduce health care costs—by $9 billion over 10 years for Medicare patients, according to a 2007 Commonwealth Fund report.

There’s also an ethical case to be made, suggests MGH’s Sepucha, who has argued that performing elective surgery on a patient who wasn’t a fully informed participant in choosing that option qualifies as a medical error. In a 2012 report for the King’s Fund, Albert Mulley, director of the Dartmouth Center for Health Care Delivery Science, and his colleagues asserted that misdiagnosing patients’ preferences may be less obvious than misdiagnosing disease, but the consequences for the patient can be just as severe.

YET WHILE THERE MAY BE A GROWING CONSENSUS that patients should be more engaged in decisions, making that happen has been a challenge. In a paper published in Health Affairs earlier this year, Lin and her colleagues described their attempt to implement decision aids at five California clinics. Project leaders tried to persuade clinicians to use the tools in much the way pharmaceutical representatives market drugs—offering lunches, training and free coffee mugs.

Still, only about 10% of eligible patients ever got a chance to use the decision aids. And although physicians were nearly unanimous in their support for shared decision making, and a majority said they would use decision aids if they were available, it tended to be staff members, not doctors, who handed out the few that were offered.

One barrier to shared decision making seems to be a lack of understanding about what it really entails. “Clinicians tell us: ‘We do that already,’ ” says Richard Thomson, professor of epidemiology and public health at Newcastle University in the United Kingdom. Yet there’s little evidence that patients are being asked to participate in deciding what to do. According to Thomson, physicians often feel they give patients plenty of information, but they may skip over “values elicitation”—understanding what’s important to the patient.

Having the time to teach patients about their choices is another issue, Lin says. Many doctors haven’t felt they could take the time to familiarize themselves with the tools and decide how to use them. There tends to be no financial incentive for implementing shared decision making and no penalty for not using decision aids.

Still another problem, says Peter Ubel, a physician and behavioral scientist at Duke University and author of Critical Decisions: How You and Your Doctor Can Make the Right Medical Choices Together, is that programs to increase shared decision making have tended to overemphasize decision aids. “The way this has been implemented has been to inform the hell out of patients,” Ubel says. “Information is necessary, but it’s far from sufficient.”

When Ubel and his colleagues tested two prostate cancer decision aids, one was clearer and less technical than the other, and patients who saw the more accessible aid were better informed and less likely to choose aggressive treatment. “But then they saw their doctors, and the difference between the decision aids went away,” he says. Whichever aid they’d seen, patients were still heavily influenced by what their doctors wanted.

Complicating matters is the fact that some patients may have difficulty engaging in medical decisions—because of illness, impaired cognition, a lack of medical literacy or simple unwillingness. “Clinicians have to develop a range of skills that allow them to respond to the real messiness of decision making,” Thomson says.

SHARED DECISION MAKING is mentioned in the Affordable Care Act, and it’s on the agenda of the Centers for Medicare & Medicaid Services. Several states also have gotten into the act. A 2007 law in Washington State, for example, establishes shared decision making procedures as a higher standard of informed consent; doctors who meet that standard may gain extra protection against lawsuits. Policies supporting shared decision making are also showing up internationally, particularly in the United Kingdom, Australia and Canada.

Clinical guidelines from physician organizations, meanwhile, are increasingly recognizing a patient’s role in choosing treatments. The American College of Physicians, for instance, recently published guidelines on PSA screening for prostate cancer that call for patient preferences to be factored into decisions about whether to do the controversial blood test.

But such measures will have little impact without fundamental shifts in how doctors and patients see their roles. Many physicians feel that giving patients more influence could stand in the way of improving their health. “We’re taught to do our best for our patients and to apply evidence-based medicine,” says Lin, who notes that doctors who go ahead with a procedure for which a patient is clinically eligible, without finding out what the patient may actually want, tend to feel they’re doing the right thing.

And that doesn’t apply only to doctors who trained before the movement in recent decades to empower patients. When Karen Sepucha and her colleagues taught shared decision making to MGH residents they often encountered the same tensions. “These young doctors really didn’t want to show patients the ‘do nothing’ option,” she says.

It’s understandable that physicians, long accustomed to dealing with patients who refuse to eat right, exercise or take prescribed medications, wouldn’t want just to bend to their patients’ will. In a recent commentary in JAMA, bioethicists David Shaw and Bernice Elger argue that “evidence-based persuasion”—in other words, trying to convince patients to do what you know they need—is an essential and ethical component of medical practice. And surveys show that some patients indeed prefer to leave choices about their care in the hands of their doctors.

Ubel doesn’t think doctors should leave patients to come to their own conclusions. “Patients need to be able to ask advice,” he says. What needs to change is how doctors gather information from patients before making a recommendation. So even if physicians nudge their patients toward a particular choice, they do so only after they’ve asked, “What matters to you?”